AIs big leap to tiny devices opens world of possibilities. The researchers in Microsofts Redmond, Washington, lab working on the project include, from left to right, Ajay Manchepalli, Rob De. Line, Lisa Ong, Chuck Jacobs, Ofer Dekel, Saleema Amershi, Shuayb Zarar, Chris Lovett and Byron Changuion. Make it happen. is the story of how I surrendered, my fear, took the leap, and got a life. In my case, a perfectly imperfect, fulfilling life as a mama, a working. Sometimes the best place to showcase the potential of a bold, worldchanging technology is a flower garden. Take the case of Ofer Dekel, for example. He. Free and confidential sexual health services including STI testing and treatment, contraception, pregnancy testing, emergency contraception, information and advice. Purple122/v4/7f/a1/82/7fa18286-ff08-16f2-48cd-cb80c274629d/source/392x696bb.jpg]];var lpix_1=pix_1.length;var p1_0= [[817' alt='Leap Downloads' title='Leap Downloads' />Photo by Dan De. Long. Sometimes the best place to showcase the potential of a bold, world changing technology is a flower garden. Take the case of Ofer Dekel, for example. He manages the Machine Learning and Optimization group at Microsofts research lab in Redmond, Washington. Squirrels often devoured flower bulbs in his garden and seeds from his bird feeder, depriving him and his family of blooms and birdsong. To solve the problem, he trained a computer vision model to detect squirrels and deployed the code onto a Raspberry Pi 3, an inexpensive, resource constrained single board computer. The device now keeps watch over his backyard and triggers his sprinkler system whenever the vermin pounce. Every hobbyist who owns a Raspberry Pi should be able to do that, said Dekel. Today, very few of them can. Dekel, an expert in machine learning, is aiming to solve that problem. He leads a multidisciplinary team of about 3. Microsofts research labs in Redmond and Bangalore, India, that is developing a new class of machine learning software and tools to embed artificial intelligence onto bread crumb size computer processors. Early previews of the software are available for download on Git. Hub. The project is part of a paradigm shift within the technology industry that Microsoft CEO Satya Nadella recently described during his keynote address at the companys Build 2. Leap Downloads' title='Leap Downloads' />Seattle. Game Prime World Defenders Full Movies. Were moving from what is todays mobile first, cloud first world to a new world that is going to be made up of an intelligent cloud and intelligent edge, he said. Intelligent edge. Creating the intelligent edge is a step toward realizing the promise of a world populated with tiny intelligent devices at every turn embedded in our clothes, scattered around our homes and offices and deployed to perform tasks such as anomaly detection and predictive maintenance everywhere from car engines and elevators to operating rooms and oil rigs. Today, these types of devices mostly work as sensors that collect and send data to machine learning models running in the cloud. Cloudbased Legal Practice Management Software for Law Firms and Conveyancers in Australia. Take the Feature Tour or Download a Free Brochure. News and feature lists of Linux and BSD distributions. Which version to get Get the latest stable release. If you find a bug, see how to write useful bug reports. To get a stack backtrace you need to use a debug type. Free screen sharing using Screenleap. The fast, simple, and free way to share your screen instantly for online meetings, sales demos, and collaboration. All the processing requires a lot of compute, it requires a lot of storage, said Shabnam Erfani, director of business and technical operations for Microsofts research lab in Redmond. You cant fit all that hardware into a low cost embedded device. Dekel and his colleagues are aiming to do the impossible, she added. To shrink and make machine learning so much more efficient that you can actually run it on the devices. These intelligent devices are part of the so called Internet of Things, or Io. T, except that these things are intended to be smart, or intelligent, even without an Internet connection. The researchers at Microsofts India lab who are working on the project include, clockwise from left front, Manik Varma, Praneeth Netrapalli, Chirag Gupta, Prateek Jain, Yeshwanth Cherapanamjeri, Rahul Sharma, Nagarajan Natarajan and Vivek Gupta. Photo by Mahesh Bhat. The dominant paradigm is that these devices are dumb, said Manik Varma, a senior researcher with Microsoft Research India and a co leader of the project. They sense their environment and transmit their sensor readings to the cloud where all of the machine learning happens. Unfortunately, this paradigm does not address a number of critical scenarios that we think can transform the world. Pushing machine learning to edge devices reduces bandwidth constraints and eliminates concerns about network latency, which is the time it takes for data to travel to the cloud for processing and back to the device. On device machine learning also limits battery drain from constant communication with the cloud and protects privacy by keeping personal and sensitive information local, Varma noted. The researchers imagine all sorts of intelligent devices that could be created with this method, from smart soil moisture sensors deployed for precision irrigation on remote farms to brain implants that warn users of impending seizures so that they can get to a safe place and call a caregiver. If youre driving on a highway and there isnt connectivity there, you dont want the implant to stop working, said Varma. In fact, thats where you really need it the most. Top down. The team is taking top down and bottom up approaches to the challenge of deploying machine learning models onto resource constrained devices. The top down approach involves developing algorithms that compress machine learning models trained for the cloud to run efficiently on devices such as the Raspberry Pi 3 and Raspberry Pi Zero. Html5 Banner Adobe Edge there. Many of todays machine learning models are deep neural networks, a class of predictors inspired by the biology of human brains. Dekel and his colleagues use a variety of techniques to compress deep neural networks to fit on small devices. A technique called weight quantization, for example, represents each neural network parameter with only a few bits, sometimes a single bit, instead of the standard 3. Microsoft researchers are working on systems that can run machine learning algorithms on microcontrollers as small as the one being held by Ofer Dekel, a lead researcher on the project. Photo by Dan De. Long. We can cram more parameters into a smaller space and the computer can churn through it much, much faster, said Dekel. To illustrate the difference, he played a video comparing a state of the art neural network with and without quantization trained and deployed for computer vision on Raspberry Pi 3s The models are equally accurate, but the compressed version runs about 2. Early previews of these compression and training algorithms are available for download on Git. Hub. The team is also working on tools that will enable hobbyists, makers and other non machine learning experts to navigate the end to end process of collecting and cleaning data, training the models and deploying them onto their devices. Giving these powerful machine learning tools to everyday people is the democratization of AI, said Saleema Amershi, a human computer interaction researcher in the Redmond lab. If we have the technology to put the smarts on the tiny devices, but the only people who can use it are the machine learning experts, then where have we gottenAnother compression technique being investigated by the team is pruning, or sparsification, of neural networks to remove redundancies, which should result in faster evaluation times as well as the ability to deploy onto smaller computers, such as the ARM Cortex M7. From back left, Vivek Seshadri, Harsha Vardhan Simhadri, Suresh Parthasarathy and Priyan Vaithilingam are among the researchers in Microsofts India lab who are working on the project. Photo by Mahesh Bhat. Bottom up. All this compression work will only make existing machine learning models 1. To deploy machine learning onto Cortex M0s, the smallest of the ARM processors they are physically about the size of a red pepper flake and Dekel calls them computer dust the models need to be made 1,0. There is just no way to take a deep neural network, have it stay as accurate as it is today, and consume 1. You cant do it, said Dekel. So, for that, we have a longer term approach, which is to start from scratch.